CO2-Brine SCAL Modeling: A Data- And Physics-Driven Approach to Predict Relative Permeability and Uncertainty for Geological Carbon Storage

Abstract Relative permeability is a crucial parameter for characterizing multiphase flow in dynamic reservoir simulation, especially for assessing CO2 storage capacity, residual trapping, and plume movement in deep saline aquifers. However, obtaining scCO2-brine relative permeability curves under reservoir conditions is challenging due to technical and economical challenges. This study introduces a novel data- and physics-driven approach to predict CO2-brine relative permeability and its uncertainty, providing saturation functions for primary drainage in geological CO2 storage applications. This study analyzes thirty-seven (37) supercritical CO2-brine relative permeability experiments across diverse lithologies, including limestone, sandstone, dolomite, and volcanic mafic rocks. These samples, from various depositional environments, exhibit porosity ranges from 6% to 28% and permeability between 0.16 and 865 mD. Relative permeability curves are reconciled through core-flood simulations, parameterizing key flow properties such as residual saturation, Corey’s exponents, and endpoint relative permeability. These parameters are analyzed as functions of initial water saturation, porosity, and permeability. Trend models based on wettability physics and empirical SCAL data are developed to quantify uncertainty and reconstruct representative saturation functions across different rock types. The results demonstrate that flow parameters exhibit systematic trends with initial water saturation. By integrating wettability theory, reservoir physics, and empirical SCAL data, trend-based models to quantify relative permeability uncertainty and reconstruct representative saturation functions across different saturation regions and rock types were developed. This workflow ensures that the derived saturation functions align with wettability physics and rock properties, making them suitable for field scale CO2 storage modelling. For CO2 storage in saline aquifers, the flow process is inverse in comparison to hydrocarbon production, as water saturation decreases from unity to residual water saturation when the CO2 plume propagates through the formation. Thus, a dedicated SCAL model is required for CCS applications—specifically, primary drainage for CO2 plume development and imbibition for the assessment of residual trapping. By applying this methodology, saturation functions are generated that accurately represent different rock types or saturation regions. This enables improved uncertainty quantification, enhanced residual trapping estimates, and more reliable CO2 plume performance assessments. The resulting relative permeability models are well-suited for field-scale numerical reservoir simulations, supporting the optimization of geological carbon storage strategies. This study presents a novel methodology for systematically parameterizing scCO2-brine relative permeability based on trend models and analogue data. By leveraging an extensive SCAL dataset and integrating wettability-consistent models, the approach offers a predictive framework for estimating relative permeability in different rock types. This approach enhances CO2 storage modelling, facilitates uncertainty analysis, and improves residual trapping and CO2 plume performance assessments, contributing valuable insights to understanding geological carbon-dioxide storage.